English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Estimating the individual treatment effect (ITE) from observational data is acrucial research topic that holds significant value across multiple domains.How to identify hidden confounders poses a key challenge in ITE estimation.Recent studies have incorporated the structural information of social networksto tackle this challenge, achieving notable advancements. However, thesemethods utilize graph neural networks to learn the representation of hiddenconfounders in Euclidean space, disregarding two critical issues: (1) thesocial networks often exhibit a scalefree structure, while Euclidean embeddingssuffer from high distortion when used to embed such graphs, and (2) eachego-centric network within a social network manifests a treatment-relatedcharacteristic, implying significant patterns of hidden confounders. To addressthese issues, we propose a novel method called Treatment-Aware HyperbolicRepresentation Learning (TAHyper). Firstly, TAHyper employs the hyperbolicspace to encode the social networks, thereby effectively reducing thedistortion of confounder representation caused by Euclidean embeddings.Secondly, we design a treatment-aware relationship identification module thatenhances the representation of hidden confounders by identifying whether anindividual and her neighbors receive the same treatment. Extensive experimentson two benchmark datasets are conducted to demonstrate the superiority of ourmethod.

Treatment-Aware Hyperbolic Representation Learning for Causal Effect Estimation with Social Networks